The overall objective of this application is to develop and validate methods for vastly undersampled radial acquisitions and constrained reconstructions for phase contrast MR angiography of the renal arteries. Coupled with novel strategies addressing respiratory motion, we aim to reduce the total scan time necessary for a high resolution mapping of 3D velocity fields, time-resolved with the cardiac cycle, and with volumetric coverage within a clinically acceptable duration. This data set will have a spatial resolution comparable or superior to single breath-hold contrast-enhanced MR angiography, the current clinical standard in MRI. In addition to information on the vessel geometry, functional parameters derived from the velocity fields, such as flow and trans-stenotic pressure gradients are available to identify hemodynamically significant stenoses. Disadvantages of current phase contrast MRA approaches include inaccuracies and artifacts from intra-voxel dephasing and extensive scan times. These issues will be overcome by the superior spatial resolution to decrease voxel size and by the combination of a 3D radial undersampling method called VIPR (Vastly undersampled Isotropic Projection Reconstruction) with a highly constrained reconstruction method called HYPR (HighlY constrained back PRojection) and motion compensation techniques to reduce scan times. We hypothesize that the availability of information on the anatomy and trans-stenotic pressure gradients could significantly improve the non-invasive assessment of patients currently identified by MRI with a moderate renal artery stenosis (50-75%). Currently, these patients undergo invasive X-ray angiography with intraarterial pressure measurements and a significant number of patients is found to not have a hemodynamically significant stenosis and does not receive surgical treatment. PC HYPR VIPR has the potential to identify these patients in a non-invasive fashion. In addition, the injection of a contrast agent is not required for PC HYPR VIPR. Therefore, patients that would currently be excluded from standard MRA exams because of the potential development of nephrogenic systemic fibrosis (NSF) could still be diagnosed with this approach. The first specific aim is to develop the techniques necessary to extend the PC VIPR method to anatomical regions where respiratory motion is present. This will include gating and motion analysis and correction techniques. Advanced processing and visualization tools will be developed to facilitate efficient data handling and diagnosis from the large data sets. The second specific aim is the development of phase sensitive HYPR processing for the processing of velocity measurements. The third specific aim is the validation of measurements of degree of stenosis, flow, and pressure gradients in the renal arteries of an animal model. In this approach, the pressure measurements are obtained from two tandem pressure-sensitive intravascular fiberoptic sensors. The fourth specific aim is the validation of PC HYPR VIPR imaging in healthy volunteers and patients with renal artery stenosis and kidney transplants. Measurements of degree of stenosis, flow, pressure gradients, and image quality will be evaluated.